Skip to content

stable-audio-tools

An open PyTorch library for conditional audio generation

Visit official site

What stable-audio-tools is

The library is described as an open-source PyTorch codebase providing training and inference code for generative audio models.

What you can do with it

  • Run conditional audio generation
  • Run unconditional audio generation
  • Train models across multiple GPUs
  • Customise models and training settings

Example uses

  • Generate music with a latent diffusion model
  • Denoise audio with an autoencoder
  • Synthesise speech with a pretrained model

Pros & cons

✓ What we like

  • Training and inference code are both provided
  • Models and training settings can be customised

! What to watch out for

  • Work runs through PyTorch code
  • Multi-GPU training is listed as a feature

FAQ

What is included?

The material lists autoencoders, latent diffusion models and MusicGen.

Can it train on several GPUs?

Yes, multi-GPU training is listed among the features.

Last reviewed: 2026-11-15

More AI music creation tools

View all →

How we review